Please use this identifier to cite or link to this item: http://hdl.handle.net/10125/41271

A Correlation Network Model for Structural Health Monitoring and Analyzing Safety Issues in Civil Infrastructures

File SizeFormat 
paper0122.pdf1.74 MBAdobe PDFView/Open

Item Summary

Title: A Correlation Network Model for Structural Health Monitoring and Analyzing Safety Issues in Civil Infrastructures
Authors: Fuchsberger, Alexander
Ali, Hesham
Keywords: Civil Infrastructures
Correlation Network
Dynamic Graph Clustering
Prediction Model
Structural Health Monitoring
Issue Date: 04 Jan 2017
Abstract: Structural Health monitoring (SHM) is essential to analyze safety issues in civil infrastructures and bridges. With the recent advancements in sensor technology, SHM is moving from the occasional or periodic maintenance checks to continuous monitoring. While each technique, whether it is utilizing assessment or sensors, has their advantages and disadvantages, we propose a method to predict infrastructure health based on representing data streams from multiple sources into a graph model that is more scaleable, flexible and efficient than relational or unstructured databases. The proposed approach is centered on the idea of intelligently determining similarities among various structures based on population analysis that can then be visualized and carefully studied. If some “unhealthy” structures are identified through assessments or sensor readings, the model is capable of finding additional structures with similar parameters that need to be carefully inspected. This can save time, cost and effort in inspection cycles, provide increased readiness, help to prioritize inspections, and in general lead to safer, more reliable infrastructures.
Pages/Duration: 10 pages
URI/DOI: http://hdl.handle.net/10125/41271
ISBN: 978-0-9981331-0-2
DOI: 10.24251/HICSS.2017.118
Rights: Attribution-NonCommercial-NoDerivatives 4.0 International
Appears in Collections:Big Data and Analytics: Concepts, Methods, Techniques and Applications Minitrack



Items in ScholarSpace are protected by copyright, with all rights reserved, unless otherwise indicated.